Note: Descriptions are shown in the official language in which they were submitted.
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
"IMMUNE CHECKPOINT INTERVENTION" IN CANCER
FIELD OF THE INVENTION
The present invention relates to methods for identifying a subject with cancer
who is suitable
for treatment with an immune checkpoint intervention, and to methods of
treatment of such
subjects. The invention further relates to a method for predicting or
determining the
prognosis of a subject with cancer.
BACKGROUND TO THE INVENTION
Among the most promising approaches to activating therapeutic antitumour
immunity is the
blockade of immune checkpoints. Immune checkpoints are inhibitory pathways in
the
immune system that are crucial for maintaining self-tolerance and modulating
the duration
and amplitude of physiological immune responses in peripheral tissues in order
to minimize
collateral tissue damage. It is now clear that tumours co-opt certain immune-
checkpoint
pathways as a major mechanism of immune resistance, particularly against T
cells that are
specific for tumour antigens. Because many of the immune checkpoints are
initiated by
ligand¨receptor interactions, they can be readily blocked by antibodies or
modulated by
recombinant forms of ligands or receptors.
Current approaches to immune checkpoint regulation in cancers involve a level
of
guesswork and serendipity based mostly in the order these compounds have been
made
available. CTLA4, PD-1 and PDL1 were discovered and produced in this order,
and that is
how they have been administered so far. Initial trials were carried out with
CTLA-4, as this
was the first to be approved by the FDA. Subsequently, PD-1/PDL1 treatments
were
approved and used.
WO 2015/103037 provides a method for identifying a subject as likely to
respond to
treatment with an immune checkpoint modulator, based on the discovery that
cancer cells
may harbour somatic mutations that result in neoepitopes that are recognisable
by a
patient's immune system as non-self. The identification of one or more
neoepitopes in a
cancer sample may be useful for determining which cancer patients are likely
to respond
favourably to treatment with an immune checkpoint modulator.
1
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
SUMMARY OF THE INVENTION
The present inventors have made the important and surprising determination
that cancer
patients with higher numbers of clonal neo-antigens, and/or a higher ratio of
clonal:sub-
clonal neoantigens or a low sub-clonal neo-antigen fraction, are more likely
to respond to
treatment with an immune checkpoint internvention.
As demonstrated in the present examples, patients with tumours with a high
clonal neo-
antigen burden and/or a low subclonal neo-antigen burden have a better
response to
immunotherapy with checkpoint blockade (e.g. anti-PD1 therapy). This
represents an
important contribution to the art, in that it opens up the potential for
improved and more
directed treatments and preventative modalities for treating and preventing
cancer. In this
regard, therapeutic and preventative interventions can be targeted to the
individual and to
the particular context of the cancer.
Furthermore, the present inventors have found that, surprisingly, tumour cells
with high
numbers of clonal neo-antigens exhibit similar expression profiles of immune
checkpoint
molecules, that is they exhibit a common expression profile of immune
checkpoint
molecules. This is an important contribution to the art, as it has not
previously been
demonstrated that cancers of specific types exhibit particular expression
profiles of immune
checkpoint molecules. The present inventors have shown this for the first
time, and this
finding facilitates more directed approaches to treating or preventing
particular cancers.
The present inventors have also surprisingly found that patients with higher
numbers of
clonal mutations, and a higher ratio of clonal:sub-clonal mutations, have
improved prognosis.
The present invention therefore addresses a need in the art for new,
alternative and/or more
effective ways of treating and preventing cancer.
Accordingly, the present invention provides a method for identifying a subject
with cancer
who is suitable for treatment with an immune checkpoint intervention, said
method
comprising:
(i) determining the number of clonal neo-antigens in one or more cancer cells
from
said subject; and/or
(ii) determining the ratio of clonal: sub-clonal neo-antigens and/or sub-
clonal neo-
antigen fraction in more than one cancer cell from said subject; and/or
2
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
(iii) determining the expression profile of immune checkpoint molecules in
cancer
cells and/or tumour infiltrating immune cells from said subject, or tumour
type,
wherein a higher number of clonal neo-antigens, and/or a higher ratio of
clonal:sub-clonal
neo-antigens, or lower (or low) sub-clonal neo-antigen fraction, and/or
differential immune
checkpoint molecule expression in comparison to a reference sample is
indicative of
response to an immune checkpoint intervention.
In another aspect, the invention provides a method for predicting or
determining the
prognosis of a subject with cancer, the method comprising:
(i) determining the number of clonal neo-antigens in one or more cancer cells
from
said subject; and/or
(ii) determining the ratio of clonal: sub-clonal neo-antigens and/or sub-
clonal neo-
antigen fraction in more than one cancer cell from said subject,
wherein a higher number of clonal neo-antigens and/or a higher ratio of
clonal:sub-clonal
neo-antigens, or lower (or low) sub-clonal neo-antigen fraction, is indicative
of improved
prognosis.
In a further aspect, the invention provides a method of treating or preventing
cancer in a
subject, wherein said method comprises the following steps:
i) identifying a subject with cancer who is suitable for treatment with an
immune
checkpoint intervention according to the method of the invention; and
ii) treating said subject with an immune checkpoint intervention.
In a yet further aspect, the invention provides a method of treating or
preventing cancer in a
subject which comprises treating a subject with cancer with an immune
checkpoint
intervention, wherein the subject has been determined to have
(i) a higher number of clonal neo-antigens; and/or
(ii) a higher ratio of clonal:sub-clonal neo-antigens, or lower (or low) sub-
clonal neo-
antigen fraction; and/or
(iii) a differential immune checkpoint molecule expression in comparison to a
reference sample.
The invention also provides an immune checkpoint intervention for use in a
method of
treatment or prevention of cancer in a subject, the method comprising:
i) identifying a subject with cancer who is suitable for treatment with an
immune
checkpoint intervention according to the method of the invention; and
ii) treating said subject with an immune checkpoint intervention.
3
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
The invention further provides an immune checkpoint intervention for use in
the treatment or
prevention of cancer in a subject, wherein the subject has
(i) a higher number of clonal neo-antigens; and/or
(ii) a higher ratio of clonal:sub-clonal neo-antigens, or lower (or low) sub-
clonal neo-
antigen fraction; and/or
(iii) a differential immune checkpoint molecule expression in comparison to a
reference sample.
DESCRIPTION OF THE FIGURES
Figure 1: (A) Total putative neo-antigen burden in cohort of TCGA LUAD (LUng
ADenocarcinoma) tumours. Proportion of neo-antigens arising from clonal (blue)
or
subclonal (red) mutations or those of undetermined (grey) clonality is shown.
(B) Overall
survival curves for patients with tumours exhibiting high neo-antigen burden,
defined as the
upper quartile of the cohort, (n = 30) compared to remainder of cohort (n =
86) (log-rank P =
0.011), (C) high clonal neo-antigen burden, defined as the upper quartile of
the cohort, (n =
29) compared to remainder of cohort (n = 87) (log-rank P = 0.0077), and (D)
high subclonal
neo-antigen burden, defined as the upper quartile of the cohort (n = 30)
compared to
remainder of cohort (n = 86) (log-rank P = 0.12). (E) Differentially expressed
genes between
the tumours with high clonal neo-antigen burden and low clonal neo-antigen
burden, defined
as the bottom quartile of the cohort, clustered on co-expression. Clusters of
immune genes
highlighted in the text are boxed.
Figure 2: A) Phylogenetic trees for L011 and L012, with trunk and branch
lengths
proportional to number of non-silent mutations. B) Putative neo-antigens
predicted for all
missense mutations in L011. The MTFR2D326Y neo-antigen (FAFQEYDSF) is
highlighted.
C) Putative neo-antigens predicted for all missense mutations in L012. The
CHTF18 L769V
neo-antigen (LLLDIVAPK) and MYADMR3OW neo-antigen (SPMIVGSPW) are indicated.
D,
E) MHC-multimer analysis of in vitro expanded CD8+ T lymphocytes deriving from
three
tumour regions and normal tissues for L011 (D) and L012 (E). In both cases,
frequency of
CD3+CD8+ T lymphocytes reactive to mutant peptides are indicated.
Figure 3: A) MHC-multimer analysis of non-expanded CD8+ T cells from tumour
regions 1-
3, adjacent normal lung tissue and PBMCs from patient L011 (upper panel) and
L012 (lower
panel). Frequency of MHC-multimer positive cells out of the CD3+CD8+
compartment is
4
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
indicated. B) lmmunophenotype of tumour-infiltrating CD8+ T cells from patient
L011,
comparing MTFR2-reactive CD8+ T cells (MTFR2+) with MHC-multimer negative CD8+
T
cells (MTFR2-) in the same tumour region, in normal tissue and in PBMCs. Data
shown is
from tumour Region 3 and representative of all regions. Percentage of cells
expressing
CTLA-4, PD-1, LAG-3, Ki-67 and GzmB is shown. C) Co-expression of PD-1, Ki67
and
GzmB on MTFR2- reactive (MTFR2+) and non-reactive CD8+ T cells (MTFR2-) D)
Upper
panel: Multi-color IHC of primary tumour from L011 and L012. CD8 (red),
Granzyme B (blue)
and LAG-3 (brown) are shown. Lower panel: PD-L1 staining in L011 region 3
versus
adjacent normal tissue.
Figure 4: For discovery (A-C) and validation cohort (D-F), number of clonal
neo-antigens
and fraction of subclonal neo-antigens is shown for patients with a durable
clinical benefit
(DCB), or non-durable benefit (NDB). Progression free survival in tumours with
a higher
number of neo-antigens and low subclonal fraction compared to those with a
lower number
of neo-antigens or high subclonal fraction is shown for discovery (C) and
validation (F)
cohorts. G) Clonal architecture for each sequenced tumour. PFS are reported
under barplot
and those with ongoing progression-free survival are labeled with +. PD-L1 is
indicated
below barplot: Strong (+) 50% membranous staining; Weak (+/-), 1-49%
membranous
staining; Negative (-),<1% membranous staining; Unknown (?). (H) Progression
free survival
in combined tumour cohort comparing tumours with a higher number of neo-
antigens and
low subclonal fraction with those with a lower number of neo-antigens or high
subclonal
fraction. I) Clonal architecture of CA9903 tumour sample, with HERC1 mutation
highlighted
and with subclones indicated. J) Putative neo-antigens predicted for all
missense mutations
in CA9903. The HERC/P3278S neo-antigen (ASNASSAAK) is highlighted.
Figure 5: Quartile Breakdown of LUAD Survival. Overall survival curves showing
all four
quartiles comparing patients on total neo-antigen load (A), clonal neo-antigen
load (B), and
subclonal neo-antigen load (C). Associated log-rank p-values between each
quartile is given
to the right of the plots.
Figure 6: Survival by number of SNVs in LUAD. (B) Overall survival curves of
patients
harboring tumours with high SNV burden (n = 30) compared to remainder of
cohort (n = 86)
(log-rank P = 0.01), (C) high clonal SNV burden (n = 30) compared to remainder
of cohort (n
= 86) (log-rank P = 0.014), and (D) high subclonal SNV burden (n = 30)
compared to
remainder of cohort (n = 86) (log-rank P = 0.14).
5
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Figure 7: LUSC (Lung Squamous cell carcinoma) cohort summary. (A) Total
putative neo-
antigen burden of TCGA LUSCpatients. Columns coloured to show proportion of
neo-
antigens arising from clonal blue) or subclonal (red) mutations or arising
from mutations of
undetermined (grey) clonality. (B) Overall survival curves of patients with
high neo-antigen
burden (n = 30) compared to those with a low neo-antigen burden (n = 91) (log-
rank P =
0.84), (C) high clonal neo-antigen burden (n = 29) compared to those with a
low clonal neo-
antigen burden (n = 92) (log-rank P = 0.99), and (D) high subclonal neo-
antigen burden (n =
30) compared to those with a low subclonal neo-antigen burden (n = 91) (log-
rank P = 0.32).
(E) Overall survival curves of patients with high SNV burden (n = 30) compared
to remainder
of cohort (n = 90) (log-rank P = 0.52), (F) high clonal SNV burden (n = 30)
compared to
remainder of cohort (n = 91) (log-rank P = 0.89), and (G) high subclonal SNV
burden (n =
30) compared to remainder of cohort (n = 92) (log-rank P = 0.28).
Figure 8: Differential Gene Expression Analysis. Differentially expressed
genes between the
high clonal neo-antigen burden patients and remainder of cohort, clustered on
coexpression.
Figure 9: lmmunophenotype of tumour-infiltrating CD8+ T cells from patient
L012 A)
Activation and functional phenotype of tumour-infiltrating CD8+ CHTF18-
reactive (CHTF18+)
and MYADM-reactive (MYADM+) T cells versus MHC-multimer negative CD8+ T cells
in
tumour (Multimer-), normal tissue and PBMCs. Percentage of cells expressing
CTLA-4, PD-
1, LAG-3, Ki-67 and GzmB is shown. Histograms are generated from L012, region
2 and
findings representative of all tumour regions. B) Co-expression of PD-1, Ki67
and granzyme
B on tumour-infiltrating CD8+ CHTF18-reactive (CHTF18+) and MYADM-reactive
(MYADM+) T cells compared to tumour infiltrating MHC-multimer negative CD8+ T
cells
(Multimer-). C) In vitro expanded tumour-infiltrating CD8+ T cells were
stained with MHC-
multimers loaded with either mutant or wild type peptides and analyzed by flow
cytometry.
Percentage of MHC multimer positive cells of the CD3+CD8+ gate is shown. L011
(Top
panel): Expanded CD8+ T cells from tumour region 1 recognize mutant but not
wild type
MTFR2. L012 (middle panel): Expanded CD8+ T cells from tumour region 2
recognize
mutant but not wild type CHTF18. L012 (bottom panel): Expanded CD8+ T cells
from tumour
region 2 recognize both mutant and wild type MYADM. The mutation in MYADM is
on the
anchor residue, primary affecting HLA binding and not T cell recognition.
Whilst the data
suggest that T cells in this patient can recognize both mutant and wildtype
peptides (when
stabilized in our MHC-multimer system), the very low affinity of the wild type
peptide would
prevent adequate presentation in vivo. (D) Validation of BV650 and PE-Cy7 MHC-
multimer
binding to expanded tumour-infiltrating lymphocytes from L011 and L012. To
validate the
quality of the reagents used to characterize MTFR2-, MYADM- and CHTF18-
reactive T cells
6
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
in non-expanded tumour samples, we used the same reagents to stain a larger
number of
expanded TILs. Data from L011 (left panel), and L012 (right panel) show clear
and defined
populations of MTFR2-, MYADM- and CHTF18-reactive T cells in the expanded
TILs.
Figure 10: Mutational burden and clonal architecture of (A) discovery and (B)
validation
cohort tumours.
Figure 11: PD-L1 expression for two groups of tumours. PD-L1 exhibits
significantly stronger
expression in tumours harboring a high clonal neo-antigen burden and a low
subclonal neo-
antigen fraction compared to tumours harboring a low clonal neo-antigen burden
or high
subclonal neo-antigen fraction.
Figure 12: A) Number of predicted clonal mutations in the discovery cohort
tumours from
patients with a durable clinical benefit (DCB) or with non durable benefit
(NDB). B) Subclonal
fraction in tumours from patients with a DCB or NDB C) Progression free
survival in
discovery tumours with a higher number of clonal mutations and low subclonal
fraction
compared to those with a lower number of clonal mutations or high subclonal
fraction. D)
Number of predicted clonal mutations in the validation cohort tumours from
patients with a
DCB or with NDB. E) Subclonal fraction in tumours from validation patients
with a DCB or
NDB F) Progression free survival in validation tumours with a higher number of
clonal
mutations and low subclonal fraction compared to those with a lower number of
clonal
mutations or high subclonal fraction. G) Number of clonal and subclonal
mutations for each
sequenced tumour with clonal (dark shading) and subclonal (light shading)
displayed in the
barplot. Bars are shaded to indicate clinical benefit status: DCB, green; NDB,
red. PFS are
reported under the barplot and those with ongoing progression-free survival
are labelled with
+. PD-L1 is indicated below barplot: Strong (+) 50% membraneous staining; Weak
(+/-), 1-
49% membraneous staining; Negative (-),1% membraneous staining; Unknown (?),
unassessable. H) Progression free survival in combined tumour cohort comparing
tumours
with a higher number of clonal mutations and low subclonal fraction with those
with a lower
number of clonal mutations or high subclonal fraction.
DETAILED DESCRIPTION OF THE INVENTION
A "neo-antigen" is a tumour-specific antigen which arises as a consequence of
a mutation
within a cancer cell. Thus, a neo-antigen is not expressed by healthy cells in
a subject.
7
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
The neo-antigen described herein may be caused by any non-silent mutation
which alters a
protein expressed by a cancer cell compared to the non-mutated protein
expressed by a
wild-type, healthy cell. For example, the mutated protein may be a
translocation or fusion.
A "mutation" refers to a difference in a nucleotide sequence (e.g. DNA or RNA)
in a tumour
cell compared to a healthy cell from the same individual. The difference in
the nucleotide
sequence can result in the expression of a protein which is not expressed by a
healthy cell
from the same individual.
For example, the mutation may be a single nucleotide variant (SNV), multiple
nucleotide
variants, a deletion mutation, an insertion mutation, a translocation, a
missense mutation or
a splice site mutation resulting in a change in the amino acid sequence
(coding mutation).
The mutations may be identified by Exome sequencing, RNA-seq, whole genome
sequencing and/or targeted gene panel sequencing and or routine Sanger
sequencing of
single genes. Suitable methods are known in the art.
Descriptions of Exome sequencing and RNA-seq are provided by Boa et al.
(Cancer
Informatics. 2014;13(Suppl 2):67-82.) and Ares et al. (Cold Spring Harb
Protoc. 2014 Nov
3;2014(11)1139-48); respectively. Descriptions of targeted gene panel
sequencing can be
found in, for example, Kammermeier etal. (J Med Genet. 2014 Nov; 51(11):748-
55) and Yap
KL et al. (Clin Cancer Res. 2014. 20:6605). See also Meyerson et al., Nat.
Rev. Genetics,
2010 and Mardis, Annu Rev Anal Chem, 2013. Targeted gene sequencing panels are
also
commercially available (e.g. as summarised by Biocompare
((http://vvww.biocompare.com/
Editorial-Articles/161194-Build-Your-Own-Gene-Panels-with-These-Custom-NGS-
Targeting-
Tools/)).
Sequence alignment to identify nucleotide differences (e.g. SNVs) in DNA
and/or RNA from
a tumour sample compared to DNA and/or RNA from a non-tumour sample may be
performed using methods which are known in the art. For example, nucleotide
differences
compared to a reference sample may be performed using the method described by
Koboldt
et al. (Genome Res. 2012; 22: 568-576). The reference sample may be the
germline DNA
and/or RNA sequence.
CLONAL NEO-ANTIGENS
8
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
The present inventors have determined that intratumour heterogeneity (ITH) can
cause
variation between the neo-antigens expressed in different regions of a tumour
and between
different cells in a tumour. In particular, the inventors have determined
that, within a tumour,
certain neo-antigens are expressed in all regions and essentially all cells of
the tumour whilst
other neo-antigens are only expressed in a subset of tumour regions and cells.
As such, a "clonal" or "truncal" neo-antigen is a neo-antigen which is
expressed effectively
throughout a tumour and encoded within essentially every tumour cell. A "sub-
clonal" or
"branched" neo-antigen is a neo-antigen which is expressed in a subset or a
proportion of
cells or regions in a tumour.
References herein to "essentially all" are intended to encompass the majority
of tumour cells
in a subject. For example, this may comprise 60-100% of cells, e.g. 60, 61,
62, 63, 64, 65,
66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79 80, 81, 82, 83, 84, 85,
86, 87, 88, 89,
90, 91, 92, 93, 94, 95, 96, 97, 98, 99 or 100% of tumour cells in a subject.
"Present throughout a tumour", "expressed effectively throughout a tumour" and
"encoded
within essentially every tumour cell" may mean that the clonal neo-antigen is
expressed in all
regions of the tumour from which samples are analysed.
It will be appreciated that a determination that a mutation is "encoded within
essentially
every tumour cell" refers to a statistical calculation and is therefore
subject to statistical
analysis and thresholds.
Likewise, a determination that a clonal neo-antigen is "expressed effectively
throughout a
tumour" refers to a statistical calculation and is therefore subject to
statistical analysis and
thresholds.
"Expressed effectively in essentially every tumour cell or essentially all
tumour cells" may
mean that the mutation is present all tumour cells analysed in a sample, as
determined using
appropriate statistical methods.
By way of example, the cancer cell fraction (CCF), describing the proportion
of cancer cells
that harbour a mutation may be used to determine whether mutations are clonal
or
branched. For example, the cancer cell fraction may be determined by
integrating variant
allele frequencies with copy numbers and purity estimates as described by
Landau et al.
(Cell. 2013 Feb 14;152(4):714-26).
9
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
In brief, CCF values are calculated for all mutations identified within each
and every tumour
region analysed. If only one region is used (i.e. only a single sample), only
one set of CCF
values will be obtained. This will provide information as to which mutations
are present in all
tumour cells within that tumour region, and will thereby provide an indication
if the mutation
is clonal or branched. All sub clonal mutations (i.e. CCF<1) in a tumour
region are
determined as branched, whilst clonal mutations with a CCF=1 are determined to
be clonal.
As stated, determining a clonal mutation is subject to statistical analysis
and threshold. As
such, a mutation may be identified as clonal if it is determined to have a CCF
95%
confidence interval >= 0.60, for example 0.65, 0.70, 0.75, 0.80, 0.85, 0.90,
0.95, 1.00 or
>1.00. Conversely, a mutation may be identified as branched if it is
determined to have a
CCF 95% confidence interval <= 0.60, for example 0.55, 0.50, 0.45, 0.40, 0.35,
0.30, 0.25,
0.20, 0.15, 0.10. 0.05 or 0.01, in any sample analysed.
It will be appreciated that the accuracy of a method for identifying clonal
mutations is
increased by identifying clonal mutations for more than one sample isolated
from the tumour.
TUMOUR SAMPLES
Isolation of biopsies and samples from tumours is common practice in the art
and may be
performed according to any suitable method, and such methods will be known to
one skilled
in the art.
The method of this aspect may comprise, for example, determining the mutations
present in
cancer cells from one or more tumour regions isolated from a tumour. For
example, the
mutations present in a single biopsy, or alternatively, at least two, at least
three, at least four,
at least five, at least six, at least seven, at least eight, at least nine or
at least ten or more
biopsies isolated from a tumour may be determined.
The individual tumour samples may be isolated from different regions located
throughout a
tumour within a primary site or between primary and metastases or within a
metastasis or
between metastases. For example, determining the mutations present in tumours
which are
known to display morphological disparate histology in different regions may
involve
determining the mutations present in a number of individual samples isolated
from
morphologically disparate regions.
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
The sample may be a blood sample. For example, the blood sample may comprise
circulating tumour DNA, circulating tumour cells or exosomes comprising tumour
DNA.
SUBJECT SUITABLE FOR TREATMENT
The invention provides a method for identifying a subject with cancer who is
suitable for
treatment with an immune checkpoint intervention, said method comprising
determining the
number of clonal neo-antigens in one or more cancer cells from said subject,
wherein a high
number of clonal neo-antigens is indicative of response to an immune
checkpoint
intervention.
As used herein, the term "suitable for treatment" may refer to a subject who
is more likely to
respond to treatment with an immune checkpoint intervention, or who is a
candidate for
treatment with an immune checkpoint intervention. A subject suitable for
treatment may be
more likely to respond to said treatment than a subject who is determined not
to be suitable
using the present invention. A subject who is determined to be suitable for
treatment
according to the present invention may demonstrate a durable clinical benefit
(DCB), which
may be defined as a partial response or stable disease lasting for at least 6
months, in
response to treatment with an immune checkpoint intervention.
The number of clonal neo-antigens identified or predicted in the cancer cells
obtained from
the subject may be compared to one or more pre-determined thresholds. Using
such
thresholds, subjects may be stratified into categories which are indicative of
the degree of
response to treatment.
A threshold may be determined in relation to a reference cohort of cancer
patients. The
cohort may comprise 10, 25, 50, 75, 100, 150, 200, 250, 500 or more cancer
patients. The
cohort may be any cancer cohort. Alternatively the patients may all have the
relevant or
specific cancer type of the subject in question.
In one embodiment, a "high" number of clonal neo-antigens means a number
greater than
the median number of clonal neo-antigens predicted in a reference cohort of
cancer patients,
such as the minimum number of clonal neo-antigens predicted to be in the upper
quartile of
the reference cohort.
11
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
In another embodiment, a "high" number of clonal neo-antigens may be defined
as 10, 20,
30, 40, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150,
160, 170, 180,
190 or 200 or more clonal neo-antigens.
A skilled person will appreciated that references to "high" or "higher"
numbers of clonal neo-
antigens may be context specific, and could carry out the appropriate analysis
accordingly.
The invention further provides a method for identifying a subject with cancer
who is suitable
for treatment with an immune checkpoint intervention, said method comprising
determining
the ratio of clonal:sub-clonal neo-antigens and/or sub-clonal neo-antigen
fraction in more
than one cancer cell subject, wherein a high ratio of clonal:sub-clonal neo-
antigens or
lower/low sub-clonal neo-antigen fraction is indicative of to response to an
immune
checkpoint intervention.
As above, the clonal:sub-clonal ratio may be within the context of a cohort of
subjects, either
with any cancer or with the relevant/specific cancer. Accordingly, the
clonal:sub-clonal neo-
antigen ratio may be determined by applying methods discussed above to a
reference
cohort. A "high" or "higher" clonal:sub-clonal ratio may therefore correspond
to a number
greater than the median clonal:sub-clonal ratio predicted in a reference
cohort of cancer
patients, such as the minimum clonal:sub-clonal ratio predicted to be in the
upper quartile of
the reference cohort.
In another embodiment, a "high" or "higher" clonal:sub-clonal ratio means a
ratio in the range
of 3:1 to 100:1, such as a ratio of at least 3:1, 5:1, 10:1, 15:1, 20:1, 25:1,
50:1, 75:1 or 100:1.
One skilled in the art will appreciate that the values may depend on the
cohort in question.
The fraction of subclonal neo-antigens may also be defined in relation to a
reference cohort,
as discussed above. For example, a "lower" or "low" fraction of subclonal neo-
antigens may
correspond to a fraction smaller than the median fraction of subclonal neo-
antigens
predicted in a reference cohort of cancer patients, such as the maximum number
predicted
to be in the bottom quartile of the cohort.
Alternatively, one skilled in the art will appreciate that a sub-clonal neo-
antigen fraction can
be determined (for example for each patient) by dividing the number of
subclonal
neoantigens (for example that are predicted in the one or more cancer cells
from said
subject) by the number of total neoantigens (for example that are predicted in
the one or
more cancer cells from said subject).
12
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
In one embodiment, a "lower" or "low" fraction of subclonal neo-antigens may
mean a
fraction of 25% or less, such as a fraction of 20, 15, 10, 5, 3, 2 or 1% or
less.
In a preferred embodiment, the method may comprise determining both the number
of clonal
neo-antigens and the ratio of clonal:sub-clonal neo-antigens or the fraction
of of sub-clonal
neo-antigens. As shown in the Example, combining measures of both neo-antigen
burden
and neo-antigen sub-clonal fraction was able to predict sensitivity to
pembrolizumab better
than either measure alone (see Fig. 4C), and outcome could be predicted in
almost all cases
(Fig 4G-H).
According the invention provides a method method for identifying a subject
with cancer who
is suitable for treatment with an immune checkpoint intervention, said method
comprising:
(i) determining the number of clonal neo-antigens in one or more cancer cells
from
said subject; and
(ii) determining the ratio of clonal: sub-clonal neo-antigens and/or sub-
clonal neo-
antigen fraction in more than one cancer cell from said subject;
wherein a higher number of clonal neo-antigens and a higher ratio of
clonal:sub-clonal neo-
antigens, or lower (or low) sub-clonal neo-antigen fraction, is indicative of
response to an
immune checkpoint intervention.
Furthermore, the present inventors have found that, surprisingly, tumour cells
with high
numbers of clonal neo-antigens exhibit similar expression profiles of immune
checkpoint
molecules, that is they exhibit a common expression profile of immune
checkpoint
molecules. As such, approaches to identify particular immune checkpoint
molecules whose
expression is increased or decreased relative to non-cancerous cells can also
be used to
identify patients likely to respond to checkpoint blockade therapies.
Therefore, in one aspect the invention provides a method for identifying
subjects who have
cancer who are more likely to respond to immune checkpoint interventions,
comprising
determining the expression profile of immune checkpoint molecules in cancer
cells from said
subject, or tumour type.
In one aspect the method comprises determining the expression profile of
immune
checkpoint molecules in the tumour, for example by identifying differentially
expressed
genes, e.g. relative to a suitable reference sample. The reference sample in
respect of
13
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
differential immune checkpoint molecule expression may be a non-cancerous cell
or tumour,
(e.g. with low clonal neoantigen burden) or peripheral blood lymphocytes.
For example, the expression profile of the immune checkpoint molecules may be
determined
by:
(i) determining the RNA sequence of a sample isolated from the tumour;
and/or
(ii) performing a transcriptome-wide differential gene expression analysis
to
identify differential expression of immune checkpoint-related genes (e.g.
adjusted to p<0.05).
Non-cancer cell data may be used as a comparison, for example from the same
patient or
from a standard reference.
The invention further provides a method for determining the expression profile
of immune
checkpoint molecules in a particular cancer type comprising the steps of:
(i) obtaining RNA-sequencing data from the Cancer Genome Atlas (TCGA) data
portal for a cohort of patients with the cancer of interest;
(ii) obtaining Level_3 gene-level data from each patient;
(iii) inputting the raw read counts into the package DESeq2 for analysis; and
(iv) performing a transcriptome-wide differential gene expression analysis to
identify significantly differentially expressed (adjusted p < 0.05) immune
checkpoint-related genes.
The invention thus provides a method for identifying subjects who have cancer
who are
more likely to respond to immune checkpoint interventions, comprising
determining the
expression profile of immune checkpoint molecules in cancer cells from said
subject, or
tumour type, using said method.
In a preferred aspect, differentially expressed genes between tumours with
high clonal neo-
antigen burden and low clonal neo-antigen burden are identified (see e.g.
Figure 1E). Thus,
information regarding the number of clonal neo-antigens is informative and
facilitates the
combining of the two approaches, namely identifying and targeting
subjects/tumours with a
high number of clonal neo-antigens, and further investigating the gene
expression of
immune checkpoint molecules in those subjects/tumours with a high level of
clonal neo-
antigens. This facilitates a "double-pronged" therapeutic attack.
In one aspect, said differential immune expression is upregulation or high
expression of an
immune checkpoint molecule which is an inhibitory receptor or costimulatory
receptor
compared to a suitable reference sample, wherein such upregulation or high
expression is
14
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
indicative of a response to immune checkpoint interventions targeting the
immune
checkpoint molecule that has been upregulated or shown high expression.
Gene expression profiles may, for example, be determined by a method as
described in
present Example 1.
In a preferred embodiment the immune checkpoint molecule is PD-1 and/or LAG-3.
In a
particularly preferred embodiment the subject has lung cancer, preferably non
small-cell lung
cancer.
In an alternative embodiment, the immune checkpoint molecule is CTLA4.
In a preferred embodiment the cancer is lung cancer or melanoma, preferably
non small-cell
lung cancer or melanoma.
This method may also be used in combination with the previously described
methods for
identifying a subject with cancer who is likely to respond to treatment with
an immune
checkpoint intervention.
Accordingly the invention provides a method for identifying a subject with
cancer who is
suitable for treatment with an immune checkpoint intervention, said method
comprising:
(i) determining the number of clonal neo-antigens in one or more cancer cells
from
said subject; and
(ii) determining the expression profile of immune checkpoint molecules in
cancer
cells and/or tumour infiltrating immune cells from said subject, or tumour
type,
wherein a higher number of clonal neo-antigens and differential immune
checkpoint
molecule expression in comparison to a reference sample is indicative of
response to an
immune checkpoint intervention.
METHOD OF PROGNOSIS
The present inventors have made the important and surprising determination
that cancer
patients with higher numbers of clonal neo-antigens, and/or a higher ratio of
clonal:sub-
clonal neoantigens or a low sub-clonal neo-antigen fraction, have improved
prognosis.
One skilled in the art would appreciate in the context of the present
invention that subjects
with high or higher numbers of clonal neo-antigens, for example within a
cohort of subjects
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
or within a range identified using a number of different subjects or cohorts,
may have
improved survival relative to subjects with lower numbers of clonal neo-
antigens.
A reference value for the number of clonal neo-antigens could be determined
using the
following method, with a "high number" or "higher number" being anything above
that.
Said method may involve determining the number of clonal neo-antigens
predicted in a
cohort of cancer subjects and either:
(i) determining the median number of clonal neo-antigens predicted in that
cohort;
wherein that median number is the reference value; or
(ii) determining the minimum number of clonal neo-antigens predicted to be in
the
upper quartile of that cohort, wherein that minimum number is the reference
value. (See e.g.
TCGA data analysis in the present Examples.)
Such a "median number" or "minimum number to be in the upper quartile" could
be
determined in any cancer cohort per se, or alternatively in the relevant /
specific cancer
types.
Alternatively, a "high" or "higher" number of clonal neo-antigens may be
defined as 50, 55,
60, 65, 70, 75, 80, 85, 90, 95, 100, 110, 120, 130, 140, 150, 160, 170, 180,
190 or 200 or
more clonal neo-antigens.
One skilled in the art would appreciate that references to "high" or "higher"
numbers of clonal
neo-antigens may be context specific, and could carry out the appropriate
analysis
accordingly.
As such, the present invention also provides a method for predicting or
determining the
prognosis of a subject with cancer, comprising determining the number of
clonal neo-
antigens in one or more cancer cells from the subject, wherein a higher number
of clonal
neo-antigens, for example relative to a cohort as discussed above, is
indicative of improved
prognosis. In a preferred embodiment the cancer is lung cancer or melanoma,
preferably
non small-cell lung cancer or melanoma.
In an alternative embodiment the invention comprises a method for predicting
or determining
the prognosis of a subject with cancer, the method comprising determining the
clonal:sub-
clonal ratio and/or sub-clonal neo-antigen fraction in more than one cancer
cell from said
subject, wherein a higher clonal:sub-clonal ratio and/or a lower/low sub-
clonal neo-antigen
16
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
fraction, for example relative to a cohort as discussed above, is indicative
of improved
prognosis. In a preferred embodiment the cancer is melanoma or lung cancer,
preferably
melanoma or non small-cell lung cancer.
TREATMENT OF CANCER
The present invention also provides a method of treating or preventing cancer
in a subject,
wherein said method comprises the following steps:
i) identifying a subject with cancer who is suitable for treatment with an
immune
checkpoint intervention according to the method of the invention; and
ii) treating said subject with an immune checkpoint intervention.
As defined herein "treatment" refers to reducing, alleviating or eliminating
one or more
symptoms of the disease, disorder or infection which is being treated,
relative to the
symptoms prior to treatment.
"Prevention" (or prophylaxis) refers to delaying or preventing the onset of
the symptoms of
the disease, disorder or infection. Prevention may be absolute (such that no
disease occurs)
or may be effective only in some individuals or for a limited amount of time.
The term "immune checkpoint intervention" is used herein to refer to any
therapy which
interacts with or modulates an immune checkpoint molecule. For example, an
immune
checkpoint intervention may also be referred to herein as a "checkpoint
blockade therapy",
"checkpoint modulator" or "checkpoint inhibitor".
By "inhibitor" is meant any means to prevent inhibition of T cell activity by
these pathways.
This can be achieved by antibodies or molecules that block receptor ligand
interaction,
inhibitors of intracellular signalling pathways, and compounds preventing the
expression of
immune checkpoint molecules on the T cell surface.
Checkpoint inhibitors include, but are not limited to, CTLA-4 inhibitors, PD-1
inhibitors, PD-
L1 inhibitors, Lag-3 inhibitors, Tim-3 inhibitors, TIGIT inhibitors and BTLA
inhibitors, for
example. Co-stimulatory antibodies deliver positive signals through immune-
regulatory
receptors including but not limited to ICOS, CD137, CD27 OX-40 and GITR.
Examples of suitable immune checkpoint interventions include pembrolizumab,
nivolumab,
atezolizumab and ipilimumab.
17
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
As shown in Example 1 (see Figures 5 and 7), lung tumours with a high number
of clonal
neoantigens express high levels of PD-1 and Lag-3, and in keeping, T cells
reactive to clonal
neoantigens in lung cancer subjects also express high levels of PD-1 and LAG-
3. The co-
expression of PD-1 and Lag-3 in tumours with high clonal neo-antigen burden
versus low
clonal burden suggests that simultaneous targeting of both pathways may
generate maximal
benefit.
Hence, in one aspect the invention relates to co-targeting PD-1 and Lag-3
pathways, for
example in lung cancer, either by co-administration of inhibitors targeting
each pathway or
by administration of a single reagent targeting both pathways. As an example
of the latter,
bispecific antibodies are able to bind to PD-1 and Lag-3, or PD-L1 and Lag-3.
In a preferred embodiment of the present invention, the subject is a mammal,
preferably a
cat, dog, horse, donkey, sheep, pig, goat, cow, mouse, rat, rabbit or guinea
pig, but most
preferably the subject is a human.
In one aspect the method of treatment or prevention of cancer according to the
invention
comprises the step of identifying a patient in need of said treatment or
therapy.
The cancer may be selected from, for example, bladder cancer, gastric cancer,
oesophageal
cancer, breast cancer, colorectal cancer, cervical cancer, ovarian cancer,
endometrial
cancer, kidney cancer (renal cell), lung cancer (small cell, non-small cell
and mesothelioma),
brain cancer (e.g. gliomas, astrocytomas, glioblastomas), melanoma, lymphoma,
small
bowel cancers (duodenal and jejuna!), leukemia, pancreatic cancer,
hepatobiliary tumours,
germ cell cancers, prostate cancer, head and neck cancers, thyroid cancer and
sarcomas.
In a preferred embodiment of the invention the cancer is lung cancer. In a
particularly
preferred embodiment the lung cancer is non-small cell lung cancer.
In one embodiment of the invention the cancer is melanoma.
In one aspect of the invention, the subject has pre-invasive disease, or is a
subject who has
had their primary disease resected who might require or benefit from adjuvant
therapy, such
as that provided by the present invention.
18
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Treatment using the methods of the present invention may also encompass
targeting
circulating tumour cells and/or metastases derived from the tumour.
The methods and uses for treating cancer according to the present invention
may be
performed in combination with additional cancer therapies. In particular, the
immune
checkpoint interventions according to the present invention may be
administered in
combination with co-stimulatory antibodies, chemotherapy and/or radiotherapy,
targeted
therapy or monoclonal antibody therapy.
The invention will now be further described by way of Examples, which are
meant to serve to
assist one of ordinary skill in the art in carrying out the invention and are
not intended in any
way to limit the scope of the invention.
EXAMPLES
Example 1
The clinical relevance of neo-antigens and immune modulation within the
context of NSCLC
ITH, and the identity of neo-antigen-reactive tumour-infiltrating T cells was
investigated.
Materials and Methods
Description of Patient Cohorts
Samples for sequencing (L011 and L012) were obtained from patients diagnosed
with non-
small cell lung cancer (NSCLC) who underwent definitive surgical resection
prior to receiving
any form of adjuvant therapy, such as chemotherapy or radiotherapy. Informed
consent
allowing for genome sequencing had been obtained. Both samples were collected
from
University College London Hospital, London (UCLHRTB 10/H1306/42) and were
subjected
to pathology review to establish the histological subtype: one tumour was
classified with
CK7+/TTF1+ adenocarcinoma (L011) and one tumour (L012) with squamous cell
carcinoma
histology. Detailed clinical characteristics are provided in table 51.
Samples obtained from (1) reflected a patient cohort of stage IV NSCLC, and a
detailed
description of this patient cohort, including tumour processing, can be found
in
supplementary material of (1). Detailed clinical characteristics of this
cohort are provided in
table S3.
19
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Clinical efficacy analysis
Clinical efficacy analysis was performed as in (1). In brief, objective
response to
pembrolizumab was assessed by investigator-assessed immunerelated response
criteria
(irRC) by a study radiologist. As outlined in protocol, CT scans were
performed every nine
weeks. Partial and complete responses were confirmed by a repeat imaging
occurring a
minimum of 4 weeks after the initial identification of response; unconfirmed
responses were
considered stable or progressive disease dependent on results of the second CT
scan.
Durable clinical benefit (DCB) was defined as stable disease or partial
response lasting
longer than 6 months (week 27, the time of third protocol-scheduled response
assessment).
No durable benefit (NDB) was defined as progression of disease 6 months of
beginning
therapy. For patients with ongoing response to study therapy, progression-free
survival was
censored at the date of the most recent imaging evaluation. For alive
patients, overall
survival was censored at the date of last known contact. Details regarding
response for each
patient can be found in table S2.
TCGA exome data sets
Tumour samples, with mutation calls and HLA typing described below, were
obtained from
the Cancer Genome Atlas (TCGA) for a cohort of lung adenocarcinoma (LUAD, n =
124) and
lung squamous cell carcinoma (LUSC, n = 124). SNV data was obtained from
TumourPortal
(2) for the LUAD and LUSC TCGA cohorts
(http://vvwvv.tumourportal.org/tumour_types?
ttype=LUAD I LUSC). One LUAD patient, TCGA-05-4396, was excluded for having
over
7000 low quality mutations called, mostly in a C[C>G]G context. A LUSC
patient, TCGA-18-
3409, was excluded for bearing a strong UV signature, uncharacteristic of a
LUSC tumour.
Tumour Processing
For both L011 and L012 four primary tumour regions from a single tumour mass,
separated
by 1cm intervals, and adjacent normal tissue were selected by a pathologist,
documented by
photography, and snap-frozen. For the brain metastasis in L011, four tumour
regions as
determined by hematoxylin and eosin (H&E) staining, were selected by a
pathologist in the
form of formalin-fixed, paraffin-embedded (FFPE) tissue blocks. Peripheral
blood was
collected at the time of surgery from all patients and snap-frozen.
Approximately 5x5x5mm
snap-frozen tumour tissue and 500p1 of blood was used for genomic DNA
extraction, using
the DNeasy kit (Qiagen) according to manufacturer's protocol. For the FFPE
tissue, manual
blade macrodissection was used to remove tumour-rich areas of tissue from 10-
40pm
unstained slides, aand DNA was extracted from this using the DNeasy Blood and
Tissue kit
(Qiagen) DNA was quantified by Qubit (Invitrogen) and DNA integrity was
examined
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
by agarose gel eletrophoresis. Details regarding processing of validation and
discovery
cohort can be found in supplementary material of (1).
Multi-region Whole-Exome Sequencing and variant calling
L012
For each tumour region and matched germ-line from patient L012, exome capture
was
performed on 1-2 pg DNA using the IIlumina Nextera kit according to the
manufacturer's
protocol (IIlumina). Samples were paired-end multiplex sequenced on the
IIlumina HiSeq
2500 at the Advanced Sequencing Facility at the LRI, as described previously
(3, 4). Each
captured library was loaded on the IIlumina platform and paired-end sequenced
to the
desired average sequencing depth (mean across exomes = 392.75). Raw paired end
reads
(100bp) in FastQ format generated by the IIlumina pipeline were aligned to the
full hg19
genomic assembly (including unknown contigs) obtained from GATK bundle 2.8
(5), using
bwa mem (bwa-0.7.7) (6). Picard tools v1.107 was used to clean, sort and merge
files from
the same patient region and to remove duplicate reads
(http://broadinstitute.github.io/picard).
Quality control metrics were obtained using a combination of picard tools
(1.107), GATK
(2.8.1) and FastQC (0.10.1)
(http://vvwvv.bioinformatics.babraham.ac.uk/projects/fastqc/).
SAMtools mpileup (0.1.16) (7) was used to locate non-reference positions in
tumour and
germ-line samples. Bases with a phred score of <20 or reads with a mapping-
quality <20
were skipped. BAQ computation was disabled and the coefficient for downgrading
mapping
quality was set to 50. Somatic variants between tumour and matched germ-line
were
determined using VarScan2 somatic (v2.3.6) (8) utilizing the output from
SAMtools mpileup.
Default parameters were used with the exception of minimum coverage for the
germ-line
sample that was set to 10, minimum variant frequency was changed to 0.01 and
tumour
purity was set to 0.5. VarScan2 processSomatic was used to extract the somatic
variants.
The resulting SNV calls were filtered for false positives using Varscan2's
associated
fpfilter.pl script, having first run the data through bam-readcount (0.5.1).
Only INDEL calls
classed as 'high confidence' by VarScan2 processSomatic were kept for further
analysis.
All variants were manually reviewed using Integrated Genomics Viewers (IGV)
(9), and
those showing an IIlumina specific error profile (10) were removed. Remaining
variants were
sequenced on Ion Torrent PGM sequencer (Life Technologies) to a median depth
of 1513.
For this an Ion AmpliSeqTM custom panel (Life Technologies) was designed using
the
online designer (vvwvv.ampliseq.com). Multiplex PCRs were performed on DNA
from each
region according to the manufacturer's protocol. Barcoded sequencing libraries
were TM
constructed, which were sequenced with 200 bp read length on the Ion Torrent
PGM
21
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
sequencer (Life Technologies). Sequence alignment to target regions from the
hg19 genome
was performed using the lonTorrent TorrentSuiteTM software. Variants for which
the
coverage was 50 in at least one region were selected. A variant was considered
to be
present in a region if the variant frequency was 0.01 for SNVs and 0.02 for
INDELS.
Again manual review in IGV was performed and variants that passed this stage
were used
for subsequent analyses. All variants were annotated using ANNOVAR (11) and
potential
driver mutations were defined as described in (12).
L011
The sequencing and analysis of the germline, and primary tumour regions have
previously been described in (13). Sequencing of the metastatic regions was
performed by
BGI Tech following the protocols described in (13). Computational processing
of the
metastatic regions was performed using the methods described for L012 above,
with an
average median depth across the samples of 93.7. The non-silent variants were
manually
reviewed using IGV as for L012.
Variant calling from Rizvi data
BAM files representing both the germline and tumour regions from (i) 16
samples
representing the discovery cohort and 18 samples representing a validation
cohort (Rizvi
data), were obtained and converted to FASTQ format using picard tools (1.107)
SamToFastq Alignment and variant calling was performed as described for L012
above.
Clonal analysis
For TCGA samples, the clonal status of each mutation was estimated by
integrating the wild-
type and mutant allele counts, absolute major and minor copy numbers, and
tumour purity
estimates as previously described (14). For L011 and L012 clonal status of
each mutation
was estimated based on multiregion sequencing analysis. In brief, each
mutation was
classified as clonal if identified and present in each and every tumour region
sequenced
within the tumour. Conversely, any mutations not ubiquitously present in every
tumour region
was classified as subclonal.
For discovery and validation cohort tumour, encompassing data obtained from
(1), the
cancer cell fraction of each mutation was estimated by integrating the local
copy number
(obtained from ASCAT, see below), tumour purity (also obtained from ASCAT),
and variant
allele frequency. In brief, for a given mutation we first calculated the
observed mutation copy
number, nmut, describing the fraction of tumour cells carrying a given
mutation multiplied by
the number of chromosomal copies at that locus using the following formula:
22
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
nmut = VAF1[pCN1+ CNn(1- PA
p
where VAF corresponds to the variant allele frequency at the mutated base, and
p, CNt, CNn
are respectively the tumour purity, the tumour locus specific copy number, and
the normal
locus specific copy number. We then calculated the expected mutation copy
number, nchr,
using the VAF and assigning a mutation to one of the possible copy numbers
using
maximum likelihood. We also assessed whether mutation copy number could be
better
explained by subclonal copy numbers when applicable. Ultimately, this allowed
us to obtain
modified variant and reference counts for every mutation, corrected for both
copy number
and tumour purity. All mutations were then clustered using the PyClone
Dirichlet process
clustering (15). Given that copy number and purity had already been corrected,
we set
integer copy numbers to 1 and purity to 1; allowing clustering to simply group
clonal and
subclonal mutations. We ran PyClone with 10,000 iterations and a burn-in of
1000, and
default parameters. Notably, for assessing mutation clonal status, mutations
were first
further filtered to ensure reliable clustering. In brief, only mutations with
a read depth of at
least 10 in both germline and tumour were used, a Varscan2 somatic p-value
threshold of
0.01. A minimum of 5 alternate reads was required for each variant, as well as
a minimum
tumour variant allele frequency of 1%. Mutations were also filtered such that
a maximum of 2
germline reads, and 2% germline variant allele frequency was permitted.
For two tumours, ZA6965 and GR0134, reliable copy number, mutation and purity
estimations could not be extracted, rendering clonal architecture analysis
intractable and
these tumours were omitted from the analysis
Copy Number Analysis
For data obtained from (1) processed sample exome SNP and copy number data
from
paired tumour-normal was generated using VarScan2 (v2.3.6). Varscan2 copy
number was
run using default parameters with the exception of min-coverage (21221095)and
data-ratio.
The data-ratio was calculated on a per-sample basis as described in
(22300766). The output
from Varscan was processed using the ASCAT v2.3 (20837533) to provide
segmented copy
number data and cellularity and ploidy estimates for all samples based on the
exome
sequence data. The following setting was altered from its default value:
Threshold for setting
ACF to 1 was adjusted from 0.2 to 0.15 and the package was run with gamma
setting of 1.
For TCGA samples, SNP6.0 data was processed to yield copy number information,
as
described in McGranahan, 2015.
23
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Phylo genetic Tree Construction
The phylogenetic trees were built using binary presence/absence matrices built
from the
regional distribution of variants within the tumour, as described in (12). For
tumour L011, the
primary tumour data was reanalyzed using the method described for L012 and the
L011
metastatic regions, allowing for a combined tree featuring both primary and
metastatic
regions.
HLA Typing of Patient Samples
For all TCGA patients, the 4-digit HLA type was determined using POLYSOLVER
(POLYmorphic loci reSOLVER)(16). Patients L011 and L012 were serotyped and
simultaneously genotyped using Optitype (17), which produced concordant
results.
Identification of Putative Neo-antigens
Identified non-silent mutations were used to generate a comprehensive list of
peptides 9-11
amino acids in length with the mutated amino acid represented in each possible
position.
The binding affinity of every mutant peptide and its corresponding wild-type
peptide to the
patient's germline HLA alleles was predicted using netMHCpan-2.8 (18, 19).
Candidate neo-
antigens were identified as those with a predicted binding strength of < 500
nM.
TCGA Survival Analysis
Clinical data for the TCGA patients was accessed through the TCGA data portal
and
downloaded from
https://tcgadata.nci.nih.gov/tcgafiles/ftp auth/distro
ftpusers/anonymous/tumour/CANCER.T
YPE /bcr/biotab/clin/. Survival analyses were performed in R using the
survival package.
Differential Gene Expression Analysis
RNA-sequencing data was downloaded from the TCGA data portal. For each LUAD
patient,
all available level_3' gene-level data was obtained. The raw read counts were
used as
input into the R package DESeq2 for analysis. A transcriptomewide differential
gene
expression analysis was performed and significantly differentially expressed
(adjusted p <
0.05) immune related genes (listed in Table 51) were identified. These genes
were clustered
on their co-expression using the metric 142.
Isolation of tumour-infiltrating lymphocytes (TILs) for L011 and L012
Tumours were taken directly from the operating theatre to the department of
pathology
where the sample was divided into regions. Samples were subsequently minced
under
sterile conditions followed by enzymatic digestion (RPMI-1640 (Sigma) with
Liberase TL
24
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
research grade (Roche) and DNAse I (Roche)) at 37 C for 30 minutes before
mechanical
dissociation using gentleMACS (Miltenyi Biotech). Resulting single cell
suspensions were
enriched for leukocytes by passage through a Ficoll-paque (GE Healthcare)
gradient. Live
cells were counted and frozen in human AB serum (Sigma) with 10% dimethyl
sulfoxide at -
80 C before transfer to liquid nitrogen.
In-vitro expansion of tumour-infiltrating lymphocytes for L011 and L012
TILs were expanded using a rapid expansion protocol (REP) in T25 flasks
containing EX-
VIVO media (Lonza) supplemented with 10% human AB serum (Sigma), soluble anti-
CD3
(OKT3, BioXCell), 6000IU/mL recombinant human (rhIL-2, PeproTech) and 2x107
irradiated
PBMCs (30Gy) pooled from 3 allogeneic healthy donors. Fresh media containing
rhIL-2 at
3000IU/mL was added every three days as required. Following 2 weeks of
expansion, TILs
were counted, phenotyped by flow cytometry and frozen in human AB serum
(Sigma) at -
80 C before use in relevant assays or long-term storage in liquid nitrogen.
MHC multimer generation and combinatorial encoding-flow cytometry analysis
MHC-multimers holding the predicted neoepitopes were produced in-house
(Technical
University of Denmark, laboratory of SRH). Synthetic peptides were purchased
at Pepscan
Presto, NL. HLA molecules matching the HLA-expression of L011 (HLA-A1101,
A2402, and
B3501) and L012 (HLA-A1101, A2402, and B0702) were refolded with a UV-
sensitive
peptide, and exchanged to peptides of interest following UV exposure (20-23).
Briefly, HLA
complexes loaded with UV-sensitive peptide were subjected to 366-nm UV light
(CAMAG)
for one hour at 4 C in the presence of candidate neo-antigen peptide in a 384-
well plate.
Peptide-MHC multimers were generated using a total of 9 different fluorescent
streptavidin
(SA) conjugates: PE, APC, PE-Cy7, PE-CF594, Brilliant Violet (BV)421, BV510,
BV605,
BV650, Brilliant Ultraviolet (BUV)395 (BioLegend). MHC-multimers were
generated with two
different streptavidin-conjugates for each peptide-specificity to allow a
combinatorial
encoding of each antigen responsive T cells, enabling analyses for reactivity
against up to
36 different peptides in parallel (24, 25).
Identification of neo-antigen-reactive CD8+ T cells
MHC-multimer analysis was performed on in-vitro expanded CD8+ T lymphocytes
isolated
from region-specific lung cancer samples and adjacent normal lung tissue. 290
and 355
candidate mutant peptides (with predicted HLA binding affinity <500nM,
including multiple
potential peptide variations from the same missense mutation) were synthesized
and used to
screen expanded L011 and L012 TILs respectively. For staining of expanded CD8+
T
lymphocytes, samples were thawed, treated with DNAse for 10 min, washed and
stained
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
with MHC multimer panels for 15 min at 37 C. Subsequently, cells were stained
with
LIVE/DEAD@ Fixable Near-IR Dead Cell Stain Kit for 633 or 635 nm excitation
(Invitrogen,
Life Technologies), CD8-PerCP (Invitrogen, Life Technologies) and FITC coupled
antibodies
to a panel of CD4, CD14, CD16, CD19 (all from BD Pharmingen) and CD40 (AbD
Serotec)
for an additional 20 min at 4 C. Data acquisition was performed on an LSR ll
flow cytometer
(Becton Dickinson) with FACSDiva 6 software. Cutoff values for the definition
of positive
responses were 0.005`)/0 of total CD8+ cells and 10 events.
For patient L011, HLA-B3501 MTFR2-derived multimers were found to bind the
mutated
sequence FAFQEYDSF (netMHC binding score: 22) but not the wild type sequence
FAFQEDDSF (netMHC binding score: 10) (Fig 11B and D, Fig 9C). No responses
were
found against overlapping peptides AFQEYDSFEK and KFAFQEYDSF. For patient L012
HLA-A1101 CHTF18-derived multimers bound the mutated sequence LLLDIVAPK
(netMHC
binding score: 37) but not the wild type sequence: LLLDILAPK (netMHC binding
score: 41)
(Fig 11C and E, Fig 9C). No responses were found against overlapping peptides
CLLLDIVAPK and IVAPKLRPV. Finally, HLA-B0702 MYADM-derived multimers bound the
mutated sequence SPMIVGSPW (netMHC binding score: 15) as well as the wild type
sequence SPMIVGSPR (netMHC binding score: 1329). No responses were found
against
overlapping peptides SPMIVGSPWA, SPMIVGSPWAL, SPWALTQPLGL and SPWALTQPL.
MHC-multimer analysis and multi-parametric flow cytometric phenotyping of
baseline, non-expanded tumour samples for L011 and L012
Tumour samples were thawed, washed and first stained with custom-made
MHCmultimers
for 10-15 minutes at 37 C in the dark. Cells were thereafter transferred onto
wet ice and
stained for 30 minutes, in the dark, with a panel of surface antibodies used
at the
manufacturer's recommended dilution: CD8-V500, SK1 clone (BD Biosciences), PD-
1-
BV605, EH12.2H7 clone (Biolegend), CD3-BV785, OKT3 clone (Biolegend), LAG-3-
PE,
3D5223H clone (eBioscience). Cells were permeablized for 20 minutes with use
of the
intracellular fixation and permeabilization buffer set from eBioscience. An
intracellular
staining panel was applied for 30 minutes, on ice, in the dark, and consisted
of the following
antibodies used at the manufacturers recommended dilution: granzyme B-V450,
GB11 clone
(BD Biosciences), FoxP3-PerCP-Cy5.5, PCH101 clone (eBioscience), Ki67-FITC,
clone B56
(BD Biosciences)and CTLA-4 ¨ APC, L3D10 clone (Biolegend). Data acquisition
was
performed on a BD FACSAria III flow cytometer (BD Biosciences) and analysed in
Flowjo
version 10Ø8 (Tree Star Inc.).
Immunohistochemistry for L011 and L012
26
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Samples from patients L011 and L012 and reactive human tonsils were fixed in
buffered
formalin and embedded in paraffin according to conventional histological
protocols. 2-5
micrometer tissue sections from paraffin blocks were cut and transferred on
electrically
charged slides to subject to immunohistochemistry. Details of the primary used
antibodies
are listed in the below table. To establish optimal staining conditions (i.e.
antibody dilution
and incubation time, antigen retrieval protocols, suitable chromogen) each
antibody was
tested and optimized on sections of human reactive tonsil by conventional
single
immunohistochemistry using the automated platforms BenchMark Ultra
(Ventana/Roche)
and the Bond-Ill Autostainer (Leica Microsystems) according to a protocol
described
elsewhere (26, 27).
Where available, at least two distinct antibodies raised against the same
protein were
analyzed in tonsil to confirm the specificity of its staining pattern. For
multiple staining a
protocol previously described was carried out (28). For evaluation of protein
co-expression in
the cytoplasm or cell membrane, change of the single colour of the chromogen
is noted i.e.
blue and red gave rise to a purple and brown and blue to an almost black
labelling.
lmmunohistochemistry and protein reactivity patterns were assessed by TM.
Scoring of
multiple immuno-staining was performed together with AF. Approval for this
study was
obtained from the National Research Ethics Service, Research Ethics Committee
4 (REC
Reference number 09/H0715/64).
Molecule Antibody type Clone name Dilution Source
Anti-human Rabbit 5P239 1:100 Spring Biosciences
CD8 Monoclonal Inc., Pleasanton,
CA,
US
Anti-human Mouse 236A/E7 1:100 Kind gift from
FoxP3 Monoclonal Dr G Roncador,
CNIO, Madrid (Spain)
Anti-human Rabbit 5P142 1:50 Spring Biosciences
PD-L1 Monoclonal Inc., Pleasanton,
CA,
US
Anti-human Mouse 17134 1:750 LifeSpan Biosciences
LAG-3 Monoclonal Inc.,
Nottingham, UK
Anti-human Mouse 11F1 RTU Leica Microsystems
Granzyme B Monoclonal Ltd., Newcastle-upon-
Tyne, UK)
27
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Results
A large tumour neo-antigen burden may increase tumour recognition by T cells,
reducing the
potential for immune-evasion (12). In support of the clinical relevance of
tumour neo-
antigens (7), it was found that high neo-antigen load (defined as the upper
quartile of the
number of neo-antigens predicted in the cohort) was associated with longer
overall survival
times in LUAD samples with matched clinical data (n=117) when compared to
tumours in the
remaining quartiles (Fig 1B, logrank p = 0.011; Fig 5A).
To determine whether neo-antigen clonal status (the presence of a neo-antigen
in all tumour
(clonal) compared to a subset of tumour cells (subclonal)) might influence the
relationship
with survival outcome, the cancer cell fraction (proportion of cancer cells
harboring each
mutation) was calculated and each putative neo-antigen was classified as
either clonal or
subclonal (13). Tumours harboring a high number of predicted clonal neo-
antigens (defined
as the upper quartile of the cohort) were associated with longer overall
survival compared to
all other tumours in the cohort (Fig 1C, log-rank p = 0.0077; Fig 5B).
Conversely, the number
of predicted subclonal neo-antigens was not significantly associated with
overall survival (Fig
1D, log-rank p = 0.12; Fig 5C). Although neo-antigen burden was related to
mutation burden,
we observed a stronger relationship between overall survival and number of neo-
antigens
compared to number of mutations (Fig 6). These data suggest the presence of a
high
number of clonal neo-antigens in LUAD may favor effective immunosurveillance.
The LUSC
cohort had a narrower range of putative neo-antigens (Fig 7A), with a median
absolute
deviation of 50 and interquartile range of 71 and a statistically significant
association
between overall survival and neo-antigen load was not observed in this cohort
(Fig 7 B-G).
This might reflect difficulties in dissecting the clonal architecture of
tumours from single
samples (14).
Gene expression analysis revealed 27 immune-related genes differentially
expressed
between low (defined as the lower quartile of the number of clonal neo-
antigens predicted in
the cohort) and high clonal neo-antigen cohorts (Table Si). CD8A (p=0.005) and
genes
associated with antigen presentation (TAP-1 p=0.003, STAT-1 p<0.001), T cell
infiltration
(CXCL-10 p=0.005, CXCL-9 p = p<0.001) and effector T cell function (IFN-y
p<0.001,
Granzymes B p<0.001 and H p=0.008) were up-regulated in the high clonal neo-
antigen
cohort and clustered together (Fig 1E). PD-1 (p=0.02) and lymphocyte
activation gene 3
(LAG-3, p<0.001), negative regulators of T cell function (15), were also
identified in this
cluster. PD-L1 was also significantly up-regulated (p<0.001) in the high
clonal cohort,
clustering with PD-L2. When we compared the high clonal neo-antigen tumours to
all other
28
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
tumours in the cohort, PD-L1 was identified as the most significantly
differentially expressed
immune gene (Fig 8, p<0.001).
These data suggest that a high clonal neo-antigen burden is associated with
the presence
ofactivated effector T cells potentially regulated by the expression of
specific immune
checkpoint proteins (PD-1, LAG-3, PD-L1/2).
It was next addressed whether CD8+ T cells reactive to clonal neo-antigens
could be
identified in primary NSCLC tumours. Two early stage tumours, L011 and L012,
subjected to
multi-region exome sequencing (13), permitted phylogenetic analysis and
prediction of
neo-antigens within each primary tumour region (Fig 2A). L011 included a brain
metastasis,
resected 14 months following primary surgery, subjected to multi-region
sequencing. While
both tumours were derived from female smokers (>40 pack-years), their mutation
burden
and extent of heterogeneity was distinct (Fig 2A). L011, an adenocarcinoma,
exhibited a
homogenous primary tumour and metastatic dissemination to the brain (M1-M4),
likely
originating from tumour region R3 (Fig 2A). A total of 313 neo-antigens were
predicted within
the primary tumour, 88% of which were clonal, identified in every region of
the primary
tumour (Fig 2B). Conversely, L012, a squamous cell carcinoma, exhibited a low
mutation
burden and extensive heterogeneity, with 75% of the predicted neo-antigens
being subclonal
(Fig 2A, C).
MHC-multimers loaded with predicted neo-antigens were used to screen CD8+ T
cells
expanded from different tumour regions and adjacent normal lung tissue (13).
In L011, CD8+
T cells reactive to mutant MTFR2D326Y (FAFQE YDSF), a clonal mutation with
high
predicted HLA binding in wild type (10nM) and mutant (22nM) forms (Fig 2B),
were identified
in all tumour regions (2.8-4.4%) and at lower frequency in normal regions
(0.1%) (Fig 2D). In
L012, CD8+ T cells reactive to mutant CHTF18L769V (LLDI VAPK) and MYADMR3OW
(SPMIVGSP W) were identified in all tumour regions and at lower frequencies in
normal
tissue (Fig 2E). Both were clonal mutations, CHTF18 with high predicted HLA
binding
(<50nM) in mutant and wild type forms, and MYADM with lower predicted binding
in wild
type (>1000nM) compared to mutant form (<50nM) (Fig 2C).
In L011, MTFR2-reactive CD8+ T cells could also be detected in non-expanded
TILs
(tumour infiltrating lymphocytes) (Fig 3A) from all primary tumour regions
(0.79-1.35%), and
at lower frequencies in normal tissue (0.16%) and peripheral blood mononuclear
cells
(PBMCs) (0.02%). Similarly, CHTF18-reactive and MYADM-reactive CD8+ T cells
were
identified in non-expanded samples from all tumour regions in L012 (CHTF18
0.16-0.58%,
29
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
MYADM 2.25- 2.31%) and at a lower frequency in normal lung tissue (CHTF18
0.02%,
MYADM 0.17%) and PBMCs (CHTF18 0.02%, MYADM 0.01%) (Fig 3A).
Further characterization of neo-antigen-reactive T cells in non-expanded
samples was
performed by flow cytometry. Although at low levels, CTLA-4 expression was
confined to
tumour-infiltrating CD8+ T cells for both L011 and L012, with highest levels
identified on
MTFR2, CHTF18 and MYADM-reactive T cells (Fig 3B, Fig 9A). High levels of PD-1
were
expressed by >99% of MTFR2-, CHTF18- and MYADM-reactive tumour-infiltrating
CD8+ T
cells (Fig 3B, Fig 9A), whilst lower levels were observed on CD8+ MHC-multimer
negative T
cells in tumour, normal tissue and PBMCs. In L011, LAG-3 expression was higher
on all
tumour rinfiltrating CD8+ T cells, including MTFR2-reactive cells, relative to
normal tissue
and PBMCs (Fig 3B). LAG-3 expression was also observed in L012, although at
lower levels
(Fig 9A). IHC studies further supported these findings, identifying CD8+ T
cells co-
expressing LAG-3 in both L011 and L012 primary tumours (Fig 3D). Ki67 was
expressed at
higher levels on tumour infiltrating CD8+ T cells than in normal tissue or
PBMCs (Fig 3B, Fig
9A), however the fraction of proliferating cells was low for both neo-antigen-
reactive and
MHC-multimer negative cells (<25%). In contrast, granzyme B (GzmB) was
expressed at
high levels on all studied CD8+ T cell subsets. Importantly, whereas a large
proportion of
neo-antigen reactive T cells in the tumours appeared highly activated
expressing GzmB, the
majority of these cells coexpressed PD- 1 (>60%) and appeared to be under
proliferative
control based on Ki67 levels (Fig 3C, Fig 9B).
Expression of LAG-3 and PD-1 on T cells reactive to clonal neo-antigens,
together with
tumour PD-L1 expression (Fig 3D), strongly supports the immune-signatures
identified in
high clonal lung tumours (Fig 1E). These data support a potential role for
these specific
checkpoints in restricting the activity of T cells recognizing clonal neo-
antigens and future
studies targeting these checkpoints in NSCLC with high clonal neo-antigen
burden.
Next, it was explored whether the clonal status of putative neo-antigens might
be associated
with altered sensitivity to PD-1 blockade in NSCLC. Exome sequencing data from
a recent
study in which two independent NSCLC cohorts were treated with pembrolizumab
was
obtained (2)(Table S2), and the clonal architecture of each tumour was
dissected by
estimating the cancer cell fraction of each mutation (13) (Fig 10). As
previously reported (2),
neo-antigen burden was related to the clinical efficacy of pembrolizumab in
the discovery
and validation cohort, with a high neo-antigen repertoire associated with
improved outcome
(data not shown).
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
The relationship was also contingent upon the clonal architecture of each
tumour (Fig. 4A-
H). In the discovery cohort, every tumour exhibiting durable clinical benefit
(DCB, defined as
in (2) as partial response or stable disease lasting > 6 months) harbored a
high clonal neo-
antigen burden (defined as above or equal to the median number of clonal neo-
antigens in
the discovery cohort, 91) and a neo-antigen subclonal fraction lower than 5%
(Fig. 4A-B).
Conversely, every tumour exhibiting a non-durable benefit (NDB) harbored
either a low
clonal neo-antigen repertoire (<91) or high neo-antigen subclonal fraction
(>5%). Thus, in
the discovery cohort, combining both neo-antigen repertoire and neo-antigen
heterogeneity
(i.e. the ratio of clonal:sub-clonal neo-antigens or mutations) was able to
predict sensitivity to
pembrolizumab, better than either measure alone (Fig 4C).
Similarly, in the validation cohort, five of six tumours with a high clonal
neo-antigen burden
(defined as greater than or equal to the median of the validation cohort, 69)
and low
subclonal neo-antigen fraction (<5%) were associated with DCB (Fig 4D-F).
Conversely,
eight out of ten tumours with low clonal neo-antigen burdens or high neo-
antigen
heterogeneity were associated with NDB. For instance, despite a large neo-
antigen burden,
ZA6505 exhibited a nondurable clinical response, relapsing after 2 months.
ZA6505 was one
of the most heterogeneous tumours within the cohort, with over 80% of
mutations classified
as subclonal.
In summary, when the extent of neo-antigen heterogeneity and the clonal neo-
antigen
burden were considered together, outcome could be predicted in almost all
cases (Fig 4G-
H).
Moreover, in keeping with TCGA analysis (Fig 1E), we also observed greater PD-
L1
expression in tumours harboring a large clonal neo-antigen burden and low neo-
antigen
heterogeneity compared to those with a low neo-antigen load or high neo-
antigen
heterogeneity (P= 0.0017, X2- test, Fig 11). These results remained consistent
when
considering all mutations rather than class- I restricted putative neo-
antigens (Fig 12),
supporting the notion that unidentified MHC class ll restricted neo-antigens
may also play a
significant role in immune reactivity (6) and the need for refinement of neo-
antigen prediction
algorithms (16). responding to anti-PD-1 therapy. Previous analysis of
peripheral blood
lymphocytes (PBLs) from CA9903, a tumour with exceptional response to
pembrolizumab,
identified a CD8+ T cell population in autologous PBLs recognizing a predicted
neo-antigen
resulting from a HERC/P32785 mutation (ASNASSAAK) (2). Consistent with the
relevance
31
CA 02997651 2018-03-06
WO 2017/042394 PCT/EP2016/071471
of clonal neo-antigens, this mutation was likely present in 100% of cancer
cells within the
sequenced tumour (Fig 4I-J).
Supplementary Table Si: Differentially expressed immune genes between high
and low clonal neo-antigen patient groups
log2 Fold adjusted p-
Mean Change p-value value
GZMB 445.7851333 -1.78462652 1.18E-09 1.69E-07
TNFSF13 3532.313186 0.825249238 6.24E-08 4.13E-06
1L6 448.0627177 -1.816113115 1.13E-07 6.70E-06
TMEM173 3740.991274 0.924504705 7.26E-07 2.84E-05
I FNG 36.80074093 -1.756035642 8.98E-06 0.000202975
PD-L1 384.7156773 -1.42730081 1.44E-05 0.00029375
CXCL9 4559.513492 -1.511142807 1.50E-05 0.000302565
STAT1 17028.82171 -0.784356172 2.16E-05 0.000401426
LAG3 277.7770415 -1.129050218 2.51E-05 0.000454562
RORA 634.8938186 0.738052259 6.53E-05 0.000963038
PRDM1 1236.570377 -0.678822885 7.91E-05 0.001116763
TAP1 10724.07983 -0.763792013 0.000338393
0.003482612
GNLY 357.1056535 -1.047292568 0.000560104
0.005132755
CXCL10 2068.038219 -1.15715645 0.000568625
0.005180037
CD8A 788.7253219 -0.945122724 0.000589072
0.005306952
CSF3 49.20726456 -1.359216439 0.000907102
0.007361937
TBX21 76.56429476 -1.021741775 0.000939277
0.007541139
GZMH 186.2030449 -0.979136784 0.001007872
0.007952958
TAP2 4538.140019 -0.559264738 0.001188075
0.008964565
PD-L2 332.3493386 -0.771675574 0.002964397
0.017858103
PVR 2940.716051 -0.567512648 0.003336494
0.019565463
CD70 71.58250113 -1.050709857 0.003388254
0.019734624
PD1 147.129952 -0.886846385 0.003492638
0.020231348
VTCN1 822.1971729 1.238388678 0.005797085
0.029375393
CHUK 1499.467613 -0.303245758 0.006444708
0.031673895
50052 795.7802019 0.663903969 0.008580641
0.039140625
TNFRSF14 2676.212113 0.359555349 0.009918008
0.043580894
CD8B 213.6837071 -0.734726572 0.012492819
0.051665805
11_1 B 462.025209 -0.701631604 0.012892938
0.052917717
I Li 2A 29.20823765 -0.68364274 0.013729832
0.055446871
I Li 2B 14.43557004 0.783016768 0.01707779
0.065407095
1L2 2.444225976 0.986346011 0.018986691
0.070900521
CX3CL1 4506.608644 0.67426411 0.027478555
0.092915037
N052 41.22675253 -0.68521024 0.029626595
0.098142202
TNFRSF18 360.9297496 -0.729059895 0.029718057
0.098354315
KLRK1 182.5573646 -0.643726062 0.030255564
0.099453935
MADCAM1 7.273228749 -0.784565543 0.048684055
0.140368258
GZMA 578.3953638 -0.635858589 0.050652532
0.144348847
32
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
VEGFA 10928.73621 -0.451684004 0.056171355
0.155239718
PRF1 618.5783468 -0.473152277 0.058353066
0.159377477
LGALS9 5310.012377 0.422760767 0.05909457 0.160839147
1L7 207.9545877 0.496316279 0.061102723
0.164605503
PTGS2 6159.88191 -0.747090631 0.063658001
0.169174461
TNFRSF4 197.5157825 -0.379177736 0.07230099 0.184487923
CD160 25.1062924 -0.460251517 0.074003068
0.187576278
TNFRSF13B 38.64965649 0.578468093 0.077739763
0.194381934
TIGIT 239.3616733 -0.479398507 0.079715132
0.19772176
TNFRSF9 107.0383741 -0.538183729 0.088224384
0.211935421
1L8 4258.061194 -0.592100426 0.102069566
0.234512774
CD86 1011.868757 -0.311180799 0.112427315
0.250957613
I RF1 3898.872307 -0.322486714 0.116076728
0.256631346
CCL5 2614.171996 -0.472610458 0.122548084
0.266275852
CD28 215.0051414 0.33366846 0.124587781 0.26940173
CD200 390.5585743 -0.295026768
0.13116523 0.27888167
HAVCR2 1325.180665 -0.296647885 0.140180116
0.291003086
MS4A1 390.2558862 0.523928974
0.15233134 0.307579262
IL12RB1 198.3081926 -0.327226759 0.161228022
0.319922079
TGFB1 3705.083958 -0.203702547 0.164951696
0.325330473
STAT3 14213.7378 0.150804216 0.169624324 0.33151912
CXCR5 95.03997316 0.416011181 0.183763767
0.348650978
001 1858.956376 -0.407500734
0.23727483 0.413799917
CD79A 1175.710552 -0.341052525 0.263424151
0.443401388
ILI ORB 3285.337481 0.152074186 0.292309086
0.475665005
I RF5 980.613802 -0.193865677 0.294881352
0.478239401
CXCR3 271.2315304 -0.283719077
0.29632793 0.479863349
TNFSF9 294.2710962 -0.291633453 0.298080985
0.481806159
N R4A1 4233.840543 0.308559097 0.307663118
0.491861702
CD69 636.205301 0.310698903 0.313829669
0.498082357
TNFRSF13C 23.54714796 -0.27661096 0.320174619
0.504889662
CTLA4 144.1973513 -0.252213052 0.337165917
0.523370695
CD80 118.3959403 -0.219168442 0.344168801
0.531186145
VEGFB 3813.436412 0.111428746 0.348255092
0.535082065
CD276 4839.942708 -0.110880457 0.354884606
0.541381686
TNFSF4 285.0994787 -0.251740957 0.367394794
0.552571633
IL15 251.954005 -0.227790082 0.372118713
0.557431842
HLA-B 117419.6406 -0.189404462 0.392700995
0.577313935
TNFSF18 10.50871731 0.278307077 0.434127114
0.615308001
CSF2 82.00905002 0.258772851 0.456961522
0.635760218
IKBKB 3029.277618 0.123880242 0.462066231
0.639685204
HLA-E 28263.91252 0.112748395 0.463864656
0.641508215
CD3D 468.5251485 -0.215789423 0.464458271
0.642153767
EOMES 73.96153664 -0.207485464 0.471039076
0.647525273
33
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
LTA 47.75821054 0.202745276
0.48236162 0.656627117
CD244 84.89011319 -0.186732614 0.494820217
0.667103087
HLA-C 76083.94895 -0.133189426 0.505495645
0.676055645
TGFBR1 3202.195003 -0.08370058 0.506750828
0.677080529
CXCL5 944.0905877 -0.270235841 0.521955968
0.689704336
HLA-G 411.9342955 0.179225426
0.52944301 0.695460893
TGFB3 950.0375094 -0.121079367 0.553603166
0.71443979
B2M 172378.9968 0.110580042 0.561947299 0.72111769
ICAM1 22061.71601 -0.143421343 0.563916412
0.722858104
CD40 1772.030076 -0.119567048 0.600081755
0.750332383
MO 48.9483278 -0.143330001 0.611619228
0.759687642
CD3E 926.6666625 -0.132795883 0.613950098
0.761354179
HLA-F 6351.352815 0.120347259 0.628199687
0.771810162
VCAM1 2159.418805 -0.130117222 0.642682858
0.782745807
CD79B 343.535879 0.127051943
0.65365394 0.790469331
CCL2 2809.121226 0.112856687 0.682702276
0.810562013
FOXP3 194.391431 0.111548571 0.689381473
0.815575495
BTLA 62.23164316 -0.102004115 0.732037222
0.845834192
SOCS1 433.4195773 0.082676919 0.734857322 0.84806292
CD2 948.4310392 -0.086505902 0.742632845
0.85329242
CD3G 131.2700215 -0.09041107 0.743838306
0.853911324
CXCL13 1506.896218 -0.118364398 0.755758076
0.861471947
ICOS 120.5538816 -0.086753989 0.760044117
0.864093608
CXCL1 960.6911493 0.085909113 0.800874166
0.889816067
CD4 5270.88003 -0.045153828 0.813688182
0.898429619
BCL6 2893.728602 -0.030670947 0.831857609
0.909937895
IL1A 52.50184149 -0.071209064 0.835686639
0.912278655
CD19 119.1478854 -0.060259488 0.8618787
0.926697933
HLA-A 92787.6356 -0.033095882 0.871822173
0.932461535
CD38 401.6927917 0.038510446 0.899895838
0.945971695
CD27 637.0929506 -0.02998534 0.908371198
0.951425293
STAT5A 1460.392543 0.003525758 0.977657023 0.99064235
IKBKG 1236.266327 0.002038592 0.989162825
0.995429737
ARG1 1.115682528 -0.318253465 0.491866502 NA
1L21 0.790890355 -1.092615876 0.040972459 NA
1L4 0.54010078 0.784036263 0.143480734 NA
34
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
Supplementary Table 52: Detailed clinical characteristics of individual
patients
Cohort
Durable
(Discovery Age Pack- Priors Dose Scheci
Event Resp Clinical
I Study ID Val idatio Histolog. , (yea . .. SE., . Smok.
__ yea PD-L1 = ^^ (mg/4 ** PFS (m ""^"^ Bene
1 SA9755 Valid NSCLC NOS 63 F Former 36 Weak 1
10 3 18.8 0 PR DCB
2 HE3202 Disc Adeno 63 F Former 58 Strong 3
10 3 14.7 0 PR DCB
3 TU0428 Disc Adeno 66 M Current 48 Negative 0
10 3 2.1 1 POD NDB
4 Y2087 Valid Adeno
68 F Never 0 Weak 5 10 3 8.3 1 SD DCB
M4945 Valid Adeno 66 M Former 40 Unknown 3 10
2 21.1 1 PR DCB
6 RI1933 Disc Adeno 60 F Former 21 Strong 1
10 3 25.2 0 PR DCB
7 7A6505 Valid Adeno 76 F Never 0 Negative 6 10
3 1.9 1 POD NDB
8 CU9061 Valid Squam 57 M Former 39 Weak 1 2 3
6.2 1 SD NDB
9 CA9903 Disc Adeno 57 M Former 80 Strong 3 10
3 14.5 1 PR DCB
5C0899 Disc Adeno 64 F Current 25 Weak 0 10
3 14.8 1 PR DCB
11 FR9547 Valid Adeno 65 F Current 25 Strong 1 2
3 12.4 1 PR DCB
12 KA3947 Disc Adeno 64 F Former 52.5 Strong 0
10 3 8.1 1 SD DCB
13 MA7027 Disc Adeno 56 M Former 37.5 Weak 1
10 2 1.8 1 POD NDB
14 ZA6965 Valid Adeno 57 F Former 25 Strong 1 2
3 14.5 0 PR DCB
AL4602 Valid Adeno 59 M Former 34 Strong 0 10 3
16.8 0 SD DCB
16 113112852 Disc Adeno 60 M Never 0 Negative 5 10
2 3.3 1 POD NDB
17 5R070761 Valid Squam 51 F Former 2.5 Negative 4
10 2 3.4 1 POD NDB
18 DI6359 Disc Adeno 61 F Current 60 Strong 6
10 3 9.8 0 PR DCB
19 513010944 Valid Squam 68 M Never 0 Unknown 2 10
3 35.7 0 PR DCB
RH090935 Valid Adeno 78 F Former 60 Strong 0 10
3 20.9 0 PR DCB
21 5C6470 Disc Adeno 59 M Current 15 Weak 0
10 2 8.3 1 SD DCB
22 BL3403 Disc Adeno 73 F Former 43.75 We a k 1
10 2 6.5 1 SD NDB
23 GR4788 Disc Squam 59 M Current 45 Negative 0
10 2 1.9 1 POD NDB
24 DM123062 Valid Adeno 50 M Never 0 Weak 6 10
2 1.9 1 POD NDB
R7495 Valid Adeno 63 M Former 73.5 Weak 2 2
3 1.4 1 POD NDB
26 WA7899 Valid Adeno 49 M Never 0 Strong 2 10
3 1.9 1 POD NDB
27 R03338 Disc Adeno 71 M Former 20 Weak 1 10
3 2.1 1 POD NDB
28 L03793 Valid Adeno 62 F Former 6 Weak 2 2 3
3.5 1 SD NDB
29 L05004 Valid Adeno 56 F Former 8 Weak 0 10
2 6.3 1 SD NDB
GR0134 Valid Adeno 80 M Former 56 Negative 0 10 3
8.3 1 PR DCB
31 VA1330 Disc Adeno 71 F Former 0.5 Unknown 1
10 3 4.1 1 SD NDB
32 NI9507 Valid Adeno 41 F Current 2.25 Weak 1 10
3 1.9 1 POD NDB
33 AU5884 Disc Adeno 64 M Former 10 Weak 2 10
2 1.8 1 POD NDB
34 VA7859 Disc Adeno 57 F Former 3.15 Unknown 1 10
3 6.3 1 SD NDB
tt,patient number. Adeno, adenocarcinoma. Squa m, squa mous cell carcinoma.
NSCLC NOS, non small-cell lung cancer, not otherwise specified. PI
**Pembrolizuma b dosed every 2 or 3 weeks as indicated.
*PDL-1 expression. Strong, >/=50% membraneous staining; Weak, 1-49%
membraneous staining; Negative, <1% membra neous staining; Unknown,
^^^^Resp. denotes best overall response to pembrolizuma b.
^^"Event (1) or censure (0) for progression-free survival
"Prior courses of cytotoxic chemotherapy. Combination chemotherapy counted as
a single course. No patient had received prior immunotherap
"Self-reported smoking status.
DCB, durable clinical benefit beyond 6 months. NDB, no dura ble benefit. NR,
not reached 6 months follow-up.
F, Female. M, Male. P, positive. No, negative. U, unknown. Smok., Smoking
status. Pack-years, product of number of packs per day and number of
Sched., Schedule of administration in weeks. Mos, months. Resp., best overall
response.
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
References
1. N. A. Rizvi etal., Cancer immunology. Mutational landscape determines
sensitivity to PD-
1 blockade in non-small cell lung cancer. Science (New York, N. Y348, 124-128
(2015).
2. M. S. Lawrence et al., Discovery and saturation analysis of cancer genes
across 21
tumour types. Nature 505, 495-501 (2014).
3. M. Gerlinger et al., Genomic architecture and evolution of clear cell renal
cell carcinomas
defined by multiregion sequencing. Nature genetics 46, 225-233 (2014).
4. M. Gerlinger et al., Intratumour heterogeneity and branched evolution
revealed by
multiregion sequencing. The New England journal of medicine 366, 883-892
(2012).
5. A. McKenna et al., The Genome Analysis Toolkit: a MapReduce framework for
analyzing
next-generation DNA sequencing data. Genome research 20, 1297-1303 (2010).
6. H. Li, R. Durbin, Fast and accurate short read alignment with Burrows-
Wheeler transform.
Bioinformatics 25, 1754-1760 (2009).
7. H. Li etal., The Sequence Alignment/Map format and SAMtools. Bioinformatics
25, 2078-
2079 (2009).
8. D. C. Koboldt et al., VarScan 2: somatic mutation and copy number
alteration discovery in
cancer by exome sequencing. Genome research 22, 568-576 (2012).
9. J. T. Robinson etal., Integrative genomics viewer. Nature biotechnology 29,
24-26 (2011).
10. K. Nakamura etal., Sequence-specific error profile of Illumine sequencers.
Nucleic Acids
Res 39, e90 (2011).
11. K. Wang, M. Li, H. Hakonarson, ANNOVAR: functional annotation of genetic
variants
from high-throughput sequencing data. Nucleic Acids Res 38, e164 (2010).
12. N. Murugaesu et al., Tracking the genomic evolution of esophageal
adenocarcinoma
through neoadjuvant chemotherapy. Cancer discovery, (2015).
13. E. C. de Bruin et al., Spatial and temporal diversity in genomic
instability processes
defines lung cancer evolution. Science (New York, N. Y346, 251-256 (2014).
14. N. McGranahan et al., Clonal status of actionable driver events and the
timing of
mutational processes in cancer evolution. Sci Trans! Med 7, 283ra254 (2015).
15. C. L. Hodgkinson etal., Tumourigenicity and genetic profiling of
circulating tumour cells
in small-cell lung cancer. Nature medicine 20, 897-903 (2014).
16. S. A. Shukla et al., Comprehensive analysis of cancer-associated somatic
mutations in
class I HLA genes. Nature biotechnology, (In press).
17. A. Szolek et al., OptiType: precision HLA typing from next-generation
sequencing data.
Bioinformatics 30, 3310-3316 (2014).
18. I. Hoof etal., NetMHCpan, a method for MHC class I binding prediction
beyond humans.
lmmunogenetics 61, 1-13 (2009).
36
CA 02997651 2018-03-06
WO 2017/042394
PCT/EP2016/071471
19. M. Nielsen et aL, NetMHCpan, a method for quantitative predictions of
peptide binding to
any HLA-A and -6 locus protein of known sequence. PloS one 2, e796 (2007).
20. M. Toebes et aL, Design and use of conditional MHC class I ligands. Nature
medicine
12, 246-251 (2006).
21. A. H. Bakker et aL, Conditional MHC class I ligands and peptide exchange
technology
for the human MHC gene products HLA-A1, -A3, -All, and -67. Proceedings of the
National
Academy of Sciences of the United States of America 105, 3825-3830 (2008).
22. T. M. Frosig et aL, Design and validation of conditional ligands for
HLAB*08:01, HLA-
B*15:01, HLA-B*35:01, and HLA-B*44:05. Cytometry A, (2015).
23. C. X. Chang et aL, Conditional ligands for Asian HLA variants facilitate
the definition of
CD8+ T-cell responses in acute and chronic viral diseases. Eur J Immunol 43,
1109-1120
(2013).
24. S. R. Hadrup et aL, Parallel detection of antigen-specific T-cell
responses by
multidimensional encoding of MHC multimers. Nat Methods 6, 520-526 (2009).
25. R. S. Andersen et aL, Parallel detection of antigen-specific T cell
responses by
combinatorial encoding of MHC multimers. Nature protocols 7, 891-902 (2012).
26. T. Marafioti et aL, Novel markers of normal and neoplastic human
plasmacytoid dendritic
cells. Blood 111, 3778-3792 (2008).
27. A. U. Akarca et aL, BRAF V600E mutation-specific antibody, a sensitive
diagnostic
marker revealing minimal residual disease in hairy cell leukaemia. Br J
Haematol 162, 848-
851 (2013).
28. T. Marafioti et aL, Phenotype and genotype of interfollicular large B
cells, a
subpopulation of lymphocytes often with dendritic morphology. Blood 102, 2868-
2876
(2003).
All documents referred to herein are hereby incorporated by reference in their
entirety, with
special attention to the subject matter for which they are referred Various
modifications and
variations of the described methods and system of the invention will be
apparent to those
skilled in the art without departing from the scope and spirit of the
invention. Although the
invention has been described in connection with specific preferred
embodiments, it should
be understood that the invention as claimed should not be unduly limited to
such specific
embodiments. Indeed, various modifications of the described modes for carrying
out the
invention which are obvious to those skilled in molecular biology, cellular
immunology or
related fields are intended to be within the scope of the following claims.
37
